1. Field of the Invention
The present invention relates generally to the fabrication of semiconductor devices and, more particularly, to a photochemical vapor deposition process for forming an oxide layer on the surface of a selected semiconductor material or an optical substrate, in a manner which enhances the oxide deposition rate.
2. Description of the Background Art
A recently developed technique for depositing oxide layers on a given substrate uses a photochemical reaction to generate neutral (nonionized) oxygen atoms, which then react with a chosen vapor phase reactant to form the desired oxide, which deposits as a layer on the substrate. This photochemical vapor deposition process is disclosed in U.S. Pat. Nos. 4,371,587 and 4,419,385, both assigned to the present assignee, and is useful in the fabrication of various devices and structures, for depositing an insulator or passivation oxide layer on a semiconductor material, glass or plastic lens, metal layer, mirrored surface, or solar cell. The use of photochemically generated neutral oxygen atoms avoids substrate damage due to charge bombardment or radiation bombardment. In addition, such a photochemical vapor deposition process can be conducted at a low temperature which avoids substrate damage due to thermal effects. As disclosed in U.S. Pat. No. 4,371,587, the neutral oxygen atoms may be generated by the mercury photosensitized dissociation of an oxygen-containing precursor or by the direct dissociation of an oxygen-containing precursor. In the mercury-sensitized reaction, a substrate is exposed to a chosen vapor phase reactant, such as silane (SiH.sub.4), an oxygen-containing precursor, such as nitrous oxide (N.sub.2 O), and mercury vapor in the presence of radiation of a predetermined wavelength e.g. 2537 angstroms). The 2537 .ANG. radiation excites the mercury atoms in the reactant gas mixture to form mercury atoms in an excited state (Hg*), approximately 5 electron volts above normal ground state, but non-ionized. The Hg* interacts with the oxygen-containing percursor, transfers energy to the precursor, and causes it to dissociate to produce atomic oxygen. The atomic oxygen then reacts with the vapor phase reactant to form the desired oxide, such as SiO.sub.2 or SiO.
In the direct photodissociation method disclosed in U.S. Pat. No. 4,371,587, atomic oxygen is formed by the direct dissociation of the oxygen-containing precursor without the assistance of mercury sensitization. In this direct process, the substrate is exposed to a chosen vapor phase reactant, such as silane, and an oxygen-containing precursor in the presence of radiation of a predetermined wavelength sufficient to cause the direct dissociation of the oxygen-containing precursor to produce atomic oxygen. When nitrous oxide is used as the oxygen-containing precursor, radiation having a wavelength within the range of 1750 to 1950 angstroms is sufficient to cause the direct dissociation of the nitrous oxide to form atomic oxygen and nitrogen as shown in Reaction (1) below. It is convenient to use 1849 .ANG. radiation for this purpose since this is the resonance line of a low pressure mercury vapor lamp which is conventionally used as the radiation source. The atomic oxygen then reacts with the vapor phase reactant, such as silane, to form the desired oxide, such as silicon dioxide or silicon monoxide. EQU N.sub.2 O+hc/.lambda.(1750-1950 .ANG.).fwdarw.O(.sup.1 D)+N.sub.2 (1)
Where
h=Planck's constant PA1 c=speed of light PA1 .lambda.=wavelength of light
The notation of O(.sup.1 D) represents a singlet-D oxygen atom, which is a neutral oxygen atom in its first excited state.
Alternatively, the atomic oxygen may be formed by the direct photochemical dissociation of molecular oxygen as shown in Reaction (2) below or of nitrogen dioxide (NO.sub.2) as shown in Reaction (3) below or of similar known materials which are capable of being dissociated to atomic oxygen by a direct photochemical reaction. EQU O.sub.2 +hc/.lambda.(1750-1950 .ANG.).fwdarw.2O(.sup.3 P) (2) EQU NO.sub.2 +hc/.lambda.(3500-6000 .ANG.).fwdarw.O(.sup.3 P)+NO (3)
The notation O(.sup.3 P) represents a triplet-P oxygen atom, which is a neutral oxygen atom in its ground state. Since molecular oxygen can react spontaneously and uncontrollably with silane, it is necessary to inhibit this spontaneous thermal oxidation process in order to permit the formation of atomic oxygen and the reaction thereof with the silane. This inhibition is accomplished by carefully controlling the operating pressure and the ratio of reactant gases and properly diluting the molecular oxygen with nitrogen gas. Higher deposition rates may be achieved using molecular oxygen rather than N.sub.2 O or NO.sub.2 as discussed above. However, in the case of oxygen, the quality of the deposited film is degraded, and unwanted powder or particulates form throughout the deposition equipment.
The mercury-sensitized photochemical vapor deposition process for depositing oxide layers is advantageous because of its higher deposition rates as compared to the direct photochemical vapor deposition process. However, in certain situations, the presence of mercury vapor may be undesirable. For example, a mercury-free environment may be critical to certain semiconductor surface passivations and dielectric bulk properties in order to avoid the incorporation of mercury into the dielectric layer or semiconductor surface and the resulting degradation of the electrical properties of the device. In addition, due to health, safety, and environmental considerations it is desirable to avoid the use of mercury. The direct photolysis process previously described avoids the problems due to mercury contamination. However, the known direct photolysis process does not have sufficiently high deposition rates to make such a process of practical use in the fabrication of semiconductor devices.
Thus, there exists in the industry the need for a photochemical vapor deposition process which avoids the use of mercury vapor, while at the same time having sufficiently high deposition rates as to be practicable.